Project Summary Ovarian cancer is the most lethal of the gynecological malignancies. Certain ovarian tumors can be particularly difficult to treat as they are often not responsive or become resistant to the chemotherapeutics that are currently in use for the treatment of ovarian cancer. Recently, it was found that clear cell and endometrioid ovarian cancer have mutations in the ARID1A gene and lack ARID1A protein expression. These mutations were found in tumors, but not in normal tissue from the same patient, suggesting a causal link between loss of ARID1A protein and tumor initiation or growth. This breakthrough, along with other sequencing efforts to profile and categorize ovarian cancer by genotype, could provide the first step in tailoring treatment for more effective survival outcomes. Paradoxically, loss of ARID1A causes growth arrest and cell death, not the uncontrolled cell division that occurs in cancer. Thus, there must be mutations in other proteins that cooperate with mutations in ARID1A to drive tumor formation. Recent data suggests that mutations in ARID1A are most often paired with activating mutations in PIK3CA, the catalytic subunit of Phosphatidylinositol 3-Kinase (PI3K), which uses ATP to convert phosphatidylinositol 4,5-bisphosphate (PIP2) to phosphatidylinositol 3,4,5-trisphosphate (PIP3). Small molecule inhibitors of PI3K and downstream components of the PI3K pathway have been developed and are currently in use as chemotherapeutics. I aim to understand how mutations in these two proteins cause cells to become transformed and whether ARID1A mutant cells from human clear cell and endometrioid ovarian cancers are more sensitive to PI3K inhibitors by virtue of this pairing. As PIK3CA mutation may be only one of the mechanisms by which cells overcome ARID1A mutation to become transformed, I plan to perform high- throughput screens for small molecules that are specifically cytotoxic to human ARID1A mutant ovarian cancer cell lines. I then hope to determine how such inhibitors debilitate ARID1A mutant cells with the aim of providing better, more personalized chemotherapeutic treatment. In summary, such studies will validate the use of ARID1A loss as a biomarker in the diagnosis of ovarian cancer and provide mechanistic insight and potentially new therapeutics for the treatment of clear cell and endometrioid ovarian cancer.